Visually and Phonologically Similar Characters in Incorrect
Simplified Chinese Words
Chao-Lin Liu†
Min-Hua Lai‡ Yi-Hsuan ChuangĹ Chia-Ying Leeľ
Department of Computer Science; †ľCenter for Mind, Brain, and Learning
National Chengchi University
ľ
Institute of Linguistics, Academia Sinica
†‡Ĺ
{†chaolin, ‡g9523, Ĺg9804}@cs.nccu.edu.tw, ľ[email protected]
Abstract
Visually and phonologically similar characters are major contributing factors for
errors in Chinese text. By defining appropriate similarity measures that consider extended Cangjie codes, we can identify visually similar characters within a
fraction of a second. Relying on the pronunciation information noted for individual characters in Chinese lexicons, we
can compute a list of characters that are
phonologically similar to a given character. We collected 621 incorrect Chinese
words reported on the Internet, and analyzed the causes of these errors. 83% of
these errors were related to phonological
similarity, and 48% of them were related
to visual similarity between the involved
characters. Generating the lists of phonologically and visually similar characters,
our programs were able to contain more
than 90% of the incorrect characters in
the reported errors.
1
Introduction
In this paper, we report the experience of our
studying the errors in simplified Chinese words.
Chinese words consist of individual characters.
Some words contain just one character, but most
words comprise two or more characters. For instance, “䟡” (mai4)1 has just one character, and
“俟‫( ”ق‬yu3 yan2) is formed by two characters.
Two most common causes for writing or typing
incorrect Chinese words are due to visual and
phonological similarity between the correct and
1
We show simplified Chinese characters followed by
their Hanyu pinyin. The digit that follows the symbols
for the sound is the tone for the character.
the incorrect characters. For instance, one might
use “ӊ” (hwa2) in the place of “㣣”(hwa4) in
“‫څ‬㣣‫׎‬ຝ” (ke1 hwa4 xing2 xiang4) partially
because of phonological similarity; one might
replace “‫( ”ܟ‬zhuo2) in “Ј㞇Κ‫( ”ܟ‬xin1 lao2
li4 zhuo2) with “䶅” (chu4) partially due to visual similarity. (We do not claim that the visual or
phonological similarity alone can explain the
observed errors.)
Similar characters are important for understanding the errors in both traditional and simplified Chinese. Liu et al. (2009a-c) applied techniques for manipulating correctness of Chinese
words to computer assisted test-item generation.
Research in psycholinguistics has shown that the
number of neighbor characters influences the
timing of activating the mental lexicon during the
process of understanding Chinese text (Kuo et al.
2004; Lee et al. 2006). Having a way to compute
and find similar characters will facilitate the
process of finding neighbor words, so can be instrumental for related studies in psycholinguistics.
Algorithms for optical character recognition for
Chinese and for recognizing written Chinese try
to guess the input characters based on sets of
confusing sets (Fan et al. 1995; Liu et al., 2004).
The confusing sets happen to be hand-crafted
clusters of visually similar characters.
It is relatively easy to judge whether two characters have similar pronunciations based on
their records in a given Chinese lexicon. We will
discuss more related issues shortly.
To determine whether two characters are visually similar is not as easy. Image processing
techniques may be useful but is not perfectly
feasible, given that there are more than fifty
thousand Chinese characters (HanDict, 2010)
and that many of them are similar to each other
in special ways. Liu et al. (2008) extend the
Cangjie codes (Cangjie, 2010; Chu, 2010) to encode the layouts and details about traditional
739
Coling 2010: Poster Volume, pages 739–747,
Beijing, August 2010
Chinese characters for computing visually similar characters. Evidence observed in psycholinguistic studies offers a cognition-based support
for the design of Liu et al.’s approach (Yeh and
Li, 2002). In addition, the proposed method
proves to be effective in capturing incorrect traditional Chinese words (Liu et al., 2009a-c).
In this paper, we work on the errors in simplified Chinese words by extending the Cangjie
codes for simplified Chinese. We obtain two lists
of incorrect words that were reported on the Internet, analyze the major reasons that contribute
to the observed errors, and evaluate how the new
Cangjie codes help us spot the incorrect characters. Results of our analysis show that phonological and visual similarities contribute similar portions of errors in simplified and traditional Chinese. Experimental results also show that, we can
catch more than 90% of the reported errors.
We go over some issues about phonological
similarity in Section 2, elaborate how we extend
and apply Cangjie codes for simplified Chinese
in Section 3, present details about our experiments and observations in Section 4, and discuss
some technical issues in Section 5.
2
Phonologically Similar Characters
The pronunciation of a Chinese character involves a sound, which consists of the nucleus and
an optional onset, and a tone. In Mandarin Chinese, there are four tones. (Some researchers include the fifth tone.)
In our work, we consider four categories of
phonological similarity between two characters:
same sound and same tone (SS), same sound and
different tone (SD), similar sound and same tone
(MS), and similar sound and different tone (MD).
We rely on the information provided in a lexicon (Dict, 2010) to determine whether two characters have the same sound or the same tone.
The judgment of whether two characters have
similar sound should consider the language experience of an individual. One who live in the
southern and one who live in the northern China
may have quite different perceptions of “similar”
sound. In this work, we resort to the confusion
sets observed in a psycholinguistic study conducted at the Academic Sinica.
Some Chinese characters are heteronyms. Let
C1 and C2 be two characters that have multiple
pronunciations. If C1 and C2 share one of their
740
pronunciations, we consider that C1 and C2 belong to the SS category. This principle applies
when we consider phonological similarity in other categories.
One challenge in defining similarity between
characters is that the pronunciations of a character can depend on its context. The most common
example of tone sandhi in Chinese (Chen, 2000)
is that the first third-tone character in words
formed by two adjacent third-tone characters will
be pronounced in the second tone. At present, we
ignore the influences of context when determining whether two characters are phonologically
similar.
Although we have confined our definition of
phonological similarity to the context of the
Mandarin Chinese, it is important to note the influence of sublanguages within the Chinese language family will affect the perception of phonological similarity. Sublanguages used in different
areas in China, e.g., Shanghai, Min, and Canton
share the same written forms with the Mandarin
Chinese, but have quite different though related
pronunciation systems. Hence, people living in
different areas in China may perceive phonological similarity in very different ways. The study in
this direction is beyond the scope of the current
study.
3
Visually Similar Characters
Figure 1 shows four groups of visually similar
characters. Characters in group 1 and group 2
differ subtly at the stroke level. Characters in
group 3 share the components on their right sides.
The shared component of the characters in group
4 appears at different places within the characters.
Radicals are used in Chinese dictionaries to
organize characters, so are useful for finding visually similar characters. The characters in group
1 and group 2 belong to the radicals “Җ” and “ ”,
respectively. Notice that, although the radical for
group 2 is clear, the radical for group 1 is not
obvious because “Җ” is not a standalone component.
However, the shared components might not be
the radicals of characters. The shared components in groups 3 and 4 are not the radicals. In
Figure 1. Examples of visually similar characters
many cases, radicals are semantic components of
Chinese characters. In groups 3 and 4, the shared
components carry information about the pronunciations of the characters. Hence, those characters are listed under different radicals, though
they do look similar in some ways.
Hence, a mechanism other than just relying on
information about characters in typical lexicons
is necessary, and we will use the extended Cangjie codes for finding visually similar characters.
ID
1
2
3
4
5
6
7
8
9
CC
Җ
җ
Ҙ
ҙ
侪
侘
侓
呆
厇
Cangjie
Җ!
ύҖ!
Җύ!
ύҖύ!
ЉζΓΜ!!
Љζ΋Μ!
ЉζΜ!
ψ΋ВВ
ψ΋Јα
ID
10
11
12
13
14
15
16
17
18
CC
偂!
㟻!
᫴!
䠑!
䡿!
䟋!
䟄!
勹!
䶈!
Cangjie
ЈЉ!
ДΓЈ
ЈЉ!
НЈ
ЈЉ!
ЕЈ
αДΓ!
αДΓ!
Җα
αΓεν!
ψ΋εν!
ψ΋΋ДΓ!
ζ΋ψΓ΋!
Table 1. Examples of Cangjie codes
3.1
Cangjie Codes for Simplified Chinese
Table 1 shows the Cangjie codes for the 13
characters listed in Figure 1 and five other
characters. The “ID” column shows the
identification number for the characters, and we
will refer to the ith character by ci, where i is the
ID. The “CC” column shows the Chinese
characters, and the “Cangjie” column shows the
Cangjie codes. Each symbol in the Cangjie codes
corresponds to a key on the keyboard, e.g. “Җ”
and “ ύ ” collocate with “W” and “L”,
respectively. Information about the complete
correspondence is available on the Wikipedia2.
Using the Cangjie codes saves us from using
image processing methods to determine the degrees of similarity between characters. Take the
Cangjie codes for the characters in group 2 (c5, c6,
and c7) for example. It is possible to find that the
characters share a common component, based on
the shared substrings of the Cangjie codes, i.e.,
“Љζ”. Using the common substring (shown in
black bold) of the Cangjie codes, we may also
find the shared component “ϭ” for characters in
group 3 (c10, c11, and c12), the shared component
“䠑” in c13 and c14, the shared component “Κ” in
c15 and c16, and the shared component “ ” in c16
and c17.
Despite the perceivable advantages, these
original Cangjie codes are not good enough. In
order to maintain efficiency in inputting Chinese
characters, the Cangjie codes have been limited
to no more than five keys. Thus, users of the
Cangjie input method must familiarize themselves with the principles for simplifying the
Cangjie codes. While the simplified codes help
the input efficiency, they also introduce difficulties and ambiguities when we compare the Cang2
en.wikipedia.org/wiki/Cangjie_input_method#Keyboard_la
yout ; last visited on 22 April 2010.
741
jie codes for computing similar characters. The
prefix “ψ΋” in c16 and c17 can represent “ ”,
“吏” (e.g., c8), and “卺” (e.g., c9). Characters
whose Cangjie codes include “ψ΋” may contain any of these three components, but they do
not really look alike.
Therefore, we augment the original Cangjie
codes by using the complete Cangjie codes and
annotate each Chinese character with a layout
identification that encodes the overall contours of
the characters. This is how Liu and his colleagues (2008) did for the Cangjie codes for traditional Chinese characters, and we employ a
similar exploration for the simplified Chinese.
3.2
Augmenting the Cangjie Codes
Figure 2 shows the twelve possible layouts that
are considered for the Cangjie codes for
simplified Chinese characters. Some of the
layouts contain smaller areas, and the rectangles
show a subarea within a character. The smaller
areas are assigned IDs between one and three.
Notice that, to maintain read-ability of the
figures, not all IDs for subareas are shown in
Figure 2. An example character is provided
below each layout. From left to right and from
top to bottom, each layout is assigned an
identification number from 1 to 12. For example,
the layout ID of “㡚” is 8. “㡚” has two parts, i.e.,
“᢭” and “ҏ”.
Researchers have come up with other ways to
Figure 2. Layouts of Chinese characters
decompose individual Chinese characters. The
Chinese Document Lab at the Academia Sinica
proposed a system with 13 operators for describing the relationships among components in Chinese characters (CDL, 2010). Lee (2010b) propose more than 30 possible layouts.
The layout of a character affects how people
perceive visual similarity between characters.
For instance, c16 in Table 1 is more similar to c17
than to c18, although they share “ ”. We rely on
the expertise in Cangjie codes reported in (Lee,
2010a) to split the codes into parts.
Table 2 shows the extended codes for some
characters listed in Table 1. The “ID” column
provides links between the characters listed in
both Table 1 and Table 2. The “CC” column
shows the Chinese characters. The “LID” column
shows the identifications for the layouts of the
characters. The columns with headings “P1”,
“P2”, and “P3” show the extended Cangjie codes,
where “Pi” shows the ith part of the Cangjie
codes, as indicated in Figure 2.
We decide the extended codes for the parts
with the help of computer programs and subjective judgments. Starting from the original Cangjie codes, we can compute the most frequent substrings just like we can compute the frequencies
of n-grams in corpora (cf. Jurafsky and Martin,
2009). Computing the most common substrings
in the original codes is not a complex task because the longest original Cangjie codes contain
just five symbols.
Often, the frequent substrings are simplified
codes for popular components in Chinese characters, e.g., “ ” and “ ”. The original codes for “ ”
and “ ” are “Љψζ” and “ψΓ΋”, but they are
often simplified to “Љζ” and “ψ΋”, respectively. When simplified, “ ” have the same
Cangjie code with “ᣠ”, and “ ” have the same
Cangjie code with “卺” and “吏”.
After finding the frequent substrings, we verify whether these frequent substrings are simplified codes for meaningful components. For meaningful components, we replace the simplified
codes with complete codes. For instance the
Cangjie codes for “侪” and “侘” are extended to
include “ψ” in Table 2, where we indicate the
extended keys that did not belong to the original
Cangjie codes in boldface and with a surrounding
box. Most of the non-meaningful frequent substrings have two keys: one is the last key of a
742
ID
5
6
7
10
11
12
13
14
15
16
17
18
19
CC
侪
侘
侓
偂
㟻
᫴
䠑
䡿
䟋
䟄
勹
䶈
䦯
LID
2
2
2
10
10
10
5
9
2
2
2
3
4
P1
ψ ζ!
Љψ
Љψ
ψ ζ!
Љψ
ψ ζ!
ДΓ!
Н!
Е!
α!
Җ!
Д Γ!
αД
Γ ΋!
ψΓ
ψΓ
Γ ΋!
ζ ΋!
ζζ
Ј!
P2
ΓΜ!
΋Μ!
Μ!
Ј!
Ј!
Ј!
ДΓ!
α!
εν!
εν!
΋ДΓ!
ψΓ!
΋΋
΋ Љ!
P3
!
!
!
Љ!
Љ!
Љ!
!
ДΓ!
!
!
!
΋!
εν!
Table 2. Examples of extended Cangjie codes
part, and the other is the first key of another part.
They were by observed by coincidence.
Although most of the examples provided in
Table 2 indicate that we expand only the first
part of the Cangjie codes, it is absolutely possible
that the other parts, i.e., P2 and P3, may need to
be extended too. c19 shows such an example.
Replacing simplified codes with complete
codes not only help us avoid incorrect matches
but also help us find matches that would be
missed due to simplification of Cangjie codes.
Using just the original Cangjie codes in Table 1,
it is not easy to determine that c18 (“䶈”) in Table
1 shares a component (“ ”) with c16 and c17 (“䟄”
and “勹”). In contrast, there is a chance to find
the similarity with the extended Cangjie codes in
Table 2, given that all of the three Cangjie codes
include “ψΓ΋”.
We can see an application of the LIDs, using
“䟄”, “勹” and “䶈” as an example. Consider the
case that we want to determine which of “勹”
and “䶈” is more similar to “䟄”. Their extended
Cangjie codes will indicate that “勹” is the answer to this question for two reasons. First, “䟄”
and “勹” belong to the same type of layout; and,
second, the shared components reside at the same
area in “䟄” and “勹”.
3.3
Similarity Measures
The main differences between the original and
the extended Cangjie codes are the degrees of
details about the structures of the Chinese characters. By recovering the details that were ignored in the original codes, our programs will be
better equipped to find the similarity between
characters.
In the current study, we experiment with three
different scoring methods to measure the visual
similarity between two characters based on their
extended Cangjie codes. Two of these methods
had been tried by Liu and his colleagues’ study
for traditional Chinese characters (Liu et al.,
2009b-c). The first method, denoted SC1, considers the total number of matched keys in the
matched parts (without considering their part
IDs). Let ci denote the ith character listed in Table
2. We have SC1(c15, c16) = 2 because of the
matched “εν”. Analogously, we have SC1(c19,
c16) = 2.
The second method, denoted SC2, includes
the score of SC1 and considers the following
conditions: (1) add one point if the matched parts
locate at the same place in the characters and (2)
if the first condition is met, an extra point will be
added if the characters belong to the same layout.
Hence, we have SC2(c15, c16) =SC1(c15,
c16)+1+1=4 because (1) the matched “εν” locate at P2 in both characters and (2) c15 and c16
belong to the same layout. Assuming that c16 belongs to layout 5, than SC2(c15, c16) would become 3. In contrast, we have SC2(c19, c16)=2. No
extra weights for the matching “εν” because it
locates at different parts in the characters. The
extra weight considers the spatial influences of
the matched parts on the perception of similarity.
While splitting the extended Cangjie codes into parts allows us to tell that c15 is more similar
to c16 than to c19, it also creates a new barrier in
computing similarity scores. An example of this
problem is that SC2(c17, c18)=0. This is because
that “ψΓ΋” at P1 in c17 can match neither “ψ
Γ” at P2 nor “΋” at P3 in c18.
To alleviate this problem, we consider SC3
which computes the similarity in three steps.
First, we concatenate the parts of a Cangjie code
for a character. Then, we compute the longest
common subsequence (LCS) (cf. Cormen et al.,
2009) of the concatenated codes of the two characters being compared, and compute a Dice’s
coefficient (cf. Croft et al., 2010) as the similarity. Let X and Y denote the concatenated, extended Cangjie codes for two characters, and let
Z be the LCS of X and Y. The similarity is defined by the following equation.
743
DiceLCS
2u Z
X Y
, where S is the length of string S
(1)
We compute another Dice’s coefficient between X and Y. The formula is the similar to (1),
except that we set Z to the longest common consecutive subsequence. We call this score
DiceLCCS . Notice that DiceLCCS d DiceLCS ,
DiceLCCS d 1 , and DiceLCS d 1 . Finally, SC3 of two
characters is the sum of their SC2, 10 u DiceLCCS ,
and 5 u DiceLCS . We multiply the Dice’s coefficients with constants to make them as influential
as the SC2 component in SC3. The constants
were not scientifically chosen, but were selected
heuristically.
4
Error Analysis and Evaluation
We evaluate the effectiveness of using the phonologically and visually similar characters to
captures errors in simplified Chinese words with
two lists of reported errors that were collected
from the Internet.
4.1
Data Sources
We need two types of data for the experiments.
The information about the pronunciation and
structures of the Chinese characters help us generate lists of similar characters. We also need
reported errors so that we can evaluate whether
the similar characters catch the reported errors.
A lexicon that provides the pronunciation information about Chinese characters and a database that contains the extended Cangjie codes are
necessary for our programs to generate lists of
characters that are phonologically and visually
similar to a given character.
It is not difficult to acquire lexicons that show
standard pronunciations for Chinese characters.
As we stated in Section 2, the main problem is
that it is not easy to predict how people in different areas in China actually pronounce the characters. Hence, we can only rely on the standards
that are recorded in lexicons.
With the procedure reported in Section 3.2, we
built a database of extended Cangjie codes for
the simplified Chinese. The database was designed to contain 5401 common characters in the
BIG5 encoding, which was originally designed
for the traditional Chinese. After converting the
traditional Chinese characters to the simplified
counterparts, the database contained only 5170
different characters.
We searched the Internet for reported errors
that were collected in real-world scenarios, and
obtained two lists of errors. The first list3 came
from the entrance examinations for senior high
schools in China, and the second list4 contained
errors observed at senior high schools in China.
We used 160 and 524 errors from the first and
the second lists, respectively, and we refer to the
combined list as the Ilist. An item of reported
error contained two parts: the correct word and
the mistaken character, both of which will be
used in our experiments.
4.2
Preliminary Data Analysis
Since our programs can compare the similarity
only between characters that are included in our
lexicon, we have to exclude some reported errors
from the Ilist. As a result, we used only 621 errors in this section.
Two native speakers subjectively classified the
causes of these errors into three categories based
on whether the errors were related to phonological similarity, visual similarity, or neither. Since
the annotators did not always agree on their classifications, the final results have five interesting
categories: “P”, “V”, “N”, “D”, and “B” in Table
3. P and V indicate that the annotators agreed on
the types of errors to be related to phonological
and visual similarity, respectively. N indicates
that the annotators believed that the errors were
not due to phonological or visual similarity. D
indicates that the annotators believed that the
errors were due to phonological or visual similarity, but they did not have a consensus. B indicates the intersection of P and V.
Table 3 shows the percentages of errors in
these categories. To get 100% from the table, we
can add up P, V, N, and D, and subtract B from
the total. In reality there are errors of type N, and
Liu and his colleagues (2009b) reported this type
of errors. Errors in this category happened to be
missing in the Ilist. Based on our and Liu’s obP
V
N
D
B
Ilist
83.1
48.3
0
3.7
35.1
Table 3. Percentages of types of errors
3
www.0668edu.com/soft/4/12/95/2008/2008091357140.htm
; last visited on 22 April 2010.
4
gaozhong.kt5u.com/soft/2/38018.html; last visited on 22
April 2010.
744
servations, the percentages of phonological and
visual similarities contribute to the errors in simplified and traditional Chinese words with similar percentages.
4.3
Experimental Procedure
We design and employ the ICCEval procedure
for the evaluation task.
At step 1, given the correct word and the correct character to be intentionally replaced with
incorrect characters, we created a list of characters based on the selection criterion. We may
choose to evaluate phonologically or visually
similar characters. For a given character, ICCEval can generate characters that are in the SS, SD,
MS, and MD categories for phonologically similar characters (cf. Section 2). For visually similar
characters, ICCEval can select characters based
on SC1, SC2, and SC3 (cf. Section 3.3). In addition, ICCEval can generate a list of characters
that belong to the same radical and have the same
number of strokes with the correct character. In
the experimental results, we refer to this type of
similar characters as RS.
At step 2, for a correct word that people originally wanted to write, we replaced the correct
character with an incorrect character with the
characters that were generated at step 1, submitted the incorrect word to Google AJAX Search
Procedure ICCEval
Input:
ccr: the correct character; cwd:
the correct word; crit: the selection criterion; num: number of requested characters; rnk: the criterion to rank the incorrect
words;
Output: a list of ranked candidates
for ccr
Steps:
1. Generate a list, L, of characters for ccr with the specified
criterion, crit. When using SC1,
SC2, or SC3 to select visually
similar characters, at most num
characters will be selected.
2. For each c in L, replace ccr in
cwd with c, submit the resulting
incorrect word to Google, and
record the ENOP.
3. Rank the list of incorrect words
generated at step 2, using the
criterion specified by rnk.
4. Return the ranked list.
API, and extracted the estimated numbers of
pages (ENOP) 5 that contained the incorrect
words. In an ordinary interaction with Google, an
ENOP can be retrieved from the search results,
and it typically follows the string “Results 110 of about” on the upper part of the browser
window. Using the AJAX API, we just have to
parse the returned results with a simple method.
Larger ENOPs for incorrect words suggest
that these words are incorrect words that people
frequently used on their web pages. Hence, we
ranked the similar characters based on their
ENOPs at step 3, and return the list.
Since the reported errors contained information about the incorrect ways to write the correct
words, we could check whether the real incorrect
characters were among the similar characters that
our programs generated at step 1 (inclusion tests).
We could also check whether the actual incorrect
characters were ranked higher in the ranked lists
(ranking tests).
Take the word “‫ک‬仠ё㤓” as an example. In
the collected data, it is reported that people wrote
this word as “‫ک‬劾ё㤓”, i.e., the second character was incorrect. Hoping to capture the error,
ICCEval generated a list of possible substitutions
for “仠” at step 1. Depending on the categories
of sources of errors, ICCEval generated a list of
characters. When aiming to test the effectiveness
of visually similar characters, we could ask ICCEval to apply SC3 to generate a list of alternatives for “仠”, possibly including “劾”, “倄”,
“လ”, and other candidates. At step 2, we created
and submitted query strings “‫ک‬劾ё㤓”, “‫ک‬倄
ё㤓”, and “‫ک‬လё㤓” to obtain the ENOPs for
the candidates. If the ENOPs were, respectively,
410000, 26100, and 7940, these candidates
would be returned in the order of “劾”, “倄”, and
“လ”. As a result, the returned list contained the
actual incorrect character “劾”, and placed “劾”
on top of the ranked list.
Notice that we considered the contexts in
which the incorrect characters appeared to rank.
We did not rank the incorrect characters with just
the unigrams. In addition, although this running
example shows that we ranked the characters
directly with the ENOPs, we also ranked the list
5
According to (Croft et al., 2010), the ENOPs may not reflect the actual number of pages on the Internet.
745
of alternatives with pointwise mutual information:
Pr(C š X )
PMI C , X ,
(2)
Pr(C ) u Pr( X )
where X is the candidate character to replace the
correct character and C is the correct word excluding the correct character to be replaced. To
compute the score of replacing “仠” with “劾” in
“‫ک‬仠ё㤓”, X = “劾”, C=“‫ک‬ɍё㤓”, and (CšX)
is “‫ک‬劾ё㤓”. (ɍ!denotes a character to be replaced.) PMI is a common tool for judging collocations in natural language processing. (cf. Jurafsky and Martin, 2009).
It would demand very much computation effort to find Pr(C). Fortunately, we do not have to
consider Pr(C) because it is a common denominator for all incorrect characters. Let X1 and X2
be two competing candidates for the correct character. We can ignore Pr(C) because of the following relationship.
PMI C , X 1 t PMI C , X 2 œ
Hence, X1 prevails if
Pr(C š X 1 ) Pr(C š X 2 )
t
Pr( X 1 )
Pr( X 2 )
scoreC , X1 scoreC , X is larger.
Pr(C š X )
Pr( X )
(3)
In our work, we approximate the probabilities
used in (3) by the corresponding frequencies that
we can collect through Google, similar to the
methods that we used to collect the ENOPs.
4.4
Experimental Results: Inclusion Tests
We ran ICCEval with 621 errors in the Ilist. The
experiments were conducted for all categories of
phonological and visual similarity. When using
SS, SD, MS, MD, and RS as the selection criterion, we did not limit the number of candidate
characters. When using SC1, SC2, and SC3 as
the criterion, we limited the number candidates
to be no more than 30. We consider only words
that the native speakers have consensus over the
causes of errors. Hence, we dropped those 3.7%
of words in Table 3, and had just 598 errors. The
ENOPs were obtained during March and April
2010.
Table 4 shows the chances that the lists, genSS
SD
MS
MD Phone
82.6
29.3
1.7
1.6
97.3
SC1
SC2
SC3
RS
Visual
Ilist 78.3
71.0
87.7
1.3
90.0
Table 4. Chances of the recommended list contains the incorrect character
Ilist
erated with different crit at step 1, contained the
incorrect character in the reported errors. In the
Ilist, there were 516 and 3006 errors that were
related to phonological and visual similarity, respectively. Using the characters generated with
the SS criterion, we captured 426 out of 516
phone-related errors, so we showed 426/516 =
82.6% in the table.
Results in Table 4 show that we captured
phone-related errors more effectively than visually-similar errors. With a simple method, we can
compute the union of the characters that were
generated with the SS, SD, MS, and MD criteria.
This integrated list suggested how well we captured the errors that were related to phones, and
we show its effectiveness under “Phone”. Similarly, we integrated the lists generated by SC1,
SC2, SC3, and RS to explore the effectiveness of
finding errors that are related to visual similarity,
and the result is shown under “Visual”.
4.5
Experimental Results: Ranking Tests
To put the generated characters into work, we
wish to put the actual incorrect character high in
the ranked list. This will help the efficiency in
supporting computer assisted test-item writing.
Having short lists that contain relatively more
confusing characters may facilitate the data preparation for psycholinguistic studies.
At step 3, we ranked the candidate characters
by forming incorrect words with other characters
in the correct words as the context and submitted
the words to Google for ENOPs. The results of
ranking, shown in Table 5, indicate that we may
just offer the leading five candidates to cover the
actual incorrect characters in almost all cases.
The “Total” column shows the total number of
errors that were captured by the selection criterion. The column “Ri” shows the percentage of
all errors, due to phonological or visual similarity,
that were re-created and ranked ith at step 3 in
ICCEVAL. The row headings show the selection
criteria that were used in the experiments. For
instance, using SS as the criterion, 70.3% of actual phone-related errors were rank first, 7.4% of
the phone-related errors were ranked second, etc.
If we recommended only 5 leading incorrect cha6
The sum of 516 and 300 is larger than 598 because
some of the characters are similar both phonologically
and visually.
746
SS
SD
MS
MD
SC1
SC2
SC3
RS
Total
R1
R2 R3 R4
426 70.3
7.4 2.9 0.4
151 25.6
2.7 0.6 0.0
9
1.4
0.4 0.0 0.0
8
1.6
0.0 0.0 0.0
235 61.3 10.3 4.3 2.0
213 53.7 11.0 3.7 2.3
263 66.7 12.7 5.7 1.7
4
1.3
0.0 0.0 0.0
Table 5. Ranking the candidates
R5
0.6
0.4
0.0
0.0
0.3
0.3
0.3
0.0
racters only with SS, we would have captured the
actual incorrect characters that were phone related 81.6% (the sum of R1 to R5) of the time.
For errors that were related to visual similarity,
recommending the top five candidates with SC3
would capture the actual incorrect characters
87.1% of the time. Since we do not show the
complete distributions, the sums over the rows
are not 100%. In the current experiments, the
worst rank was 21.
We also used PMI to rank the incorrect words.
Due to page limits, we cannot show complete
details about the results. The observed distributions in ranks were not very different from those
shown in Table 5.
5
Discussion
Compared with Liu et al.’s analysis (2009b-c)
for the traditional Chinese, the proportions of
errors related to phonological factors are almost
the same, both at about 80%. The proportion of
errors related to visual factors varied, but the averages in both studies were about 48%. A larger
scale of study is needed for how traditional and
simplified characters affect the distributions of
errors. Results shown in Table 4 suggest that it is
relatively easy to capture errors related to visual
factors in simplified Chinese. Although we cannot elaborate, we note that Cangjie codes are not
good for comparing characters that have few
strokes, e.g., c1 to c4 in Table 1. In these cases,
the coding method for Wubihua input method
(Wubihua, 2010) should be applied.
Acknowledgement
This research was supported in part by the research
contract NSC-97-2221-E-004-007-MY2 from the National Science Council of Taiwan. We thank the anonymous reviewers for constructive comments. Although we are not able to respond to all the comments
in this paper, we have done so in an extended version
of this paper.
References
Cangjie. 2010. Last visited on 22 April 2010:
en.wikipedia.org/wiki/Cangjie_input_method.
CDL. 2010. Chinese document laboratory, Academia
Sinica. Last visited on 22 April, 2010;
cdp.sinica.edu.tw/cdphanzi/. (in Chinese)
Chen, Matthew. Y. 2000. Tone Sandhi: Patterns
across Chinese Dialects, (Cambridge Studies in
Linguistics 92). Cambridge University Press.
Chu, Bong-Foo. 2010. Handbook of the Fifth Generation of the Cangjie Input Method. last visited on 22
April 2010: www.cbflabs.com/book/5cjbook/. (in Chinese)
Cormen, Thomas H., Charles E. Leiserson, Ronald L.
Rivest, and Clifford Stein. 2009. Introduction to
Algorithms, third edition. MIT Press.
Croft, W. Bruce, Donald Metzler, and Trevor Strohman, 2010. Search Engines: Information Retrieval
in Practice, Pearson.
Dict.
2010.
Last visited on 22 April
www.cns11643.gov.tw/AIDB/welcome.do
2010,
Fan, Kuo-Chin, Chang-Keng Lin, and Kuo-Sen Chou.
1995. Confusion set recognition of on-line Chinese
characters by artificial intelligence technique. Pattern Recognition, 28(3):303ԟ313.
HanDict. 2010. Last visit on 22 April 2010,
www.zdic.net/appendix/f19.htm.
Jurafsky, Daniel and James H. Martin. 2009. Speech
and Language Processing, second edition, Pearson.
Kuo, Wen-Jui, Tzu-Chen Yeh, Jun-Ren Lee, Li-Fen
Chen, Po-Lei Lee, Shyan-Shiou Chen, Low-Tone
Ho, Daisy L. Hung, Ovid J.-L. Tzeng, and JenChuen Hsieh. 2004. Orthographic and phonological
processing of Chinese characters: An fMRI study.
NeuroImage, 21(4):1721ԟ1731.
Lee, Chia-Ying, Jie-Li Tsai, Hsu-Wen Huang, Daisy
L. Hung, Ovid J.-L. Tzeng. 2006. The temporal
signatures of semantic and phonological activations
for Chinese sublexical processing: An even-related
potential study. Brain Research, 1121(1):150-159.
Lee, Hsiang. 2010a. Cangjie Input Methods in 30
Days 2. Foruto. Last visited on 22 April 2010: input.foruto.com/cccls/cjzd.html.
Lee, Mu. 2010b. A quantitative study of the formation
of Chinese characters. Last visited on 22 April
2010: chinese.exponode.com/0_1.htm. (in Chinese)
747
Liu, Chao-Lin, and Jen-Hsiang Lin. 2008. Using
structural information for identifying similar Chinese characters. Proc. of the 46th Annual Meeting
of the Association for Computational Linguistics,
short papers, 93ԟ96.
Liu, Chao-Lin, Kan-Wen Tien, Yi-Hsuan Chuang,
Chih-Bin Huang, and Juei-Yu Weng. 2009a. Two
applications of lexical information to computerassisted item authoring for elementary Chinese.
Proc. of the 22nd Int’l Conf. on Industrial Engineering & Other Applications of Applied Intelligent Systems, 470ԟ480.
Liu, Chao-Lin, Kan-Wen Tien, Min-Hua Lai, YiHsuan Chuang, and Shih-Hung Wu. 2009b. Capturing errors in written Chinese words. Proc. of the
47th Annual Meeting of the Association for Computational Linguistics, short papers, 25ԟ28.
Liu, Chao-Lin, Kan-Wen Tien, Min-Hua Lai, YiHsuan Chuang, and Shih-Hung Wu. 2009c. Phonological and logographic influences on errors in
written Chinese words. Proc. of the 7th Workshop
on Asian Language Resources, the 47th Annual
Meeting of the ACL, 84ԟ91.
Liu, Cheng-Lin, Stefan Jaeger, and Masaki Nakagawa.
2004. Online recognition of Chinese characters:
The state-of-the-art. IEEE Transaction on Pattern
Analysis and Machine Intelligence, 26(2):198ԟ213.
Wubihua. 2010. Last visited on 22 April 2010:
en.wikipedia.org/wiki/Wubihua_method.
Yeh, Su-Ling, and Jing-Ling Li. 2002. Role of structure and component in judgments of visual similarity of Chinese Characters. Journal of Experimental Psychology: Human Perception and Performance, 28(4):933ԟ947.